Glucomannoproteins in the cell wall of Saccharomyces cerevisiae contain a novel type of carbohydrate side chain - 3381y

(1)

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Glucomannoproteins in the cell wall of Saccharomyces cerevisiae contain a

novel type of carbohydrate side chain

Montijn, R.C.; van Rinsum, J.; van Schagen, F.A.; Klis, F.M.

Publication date

1994

Published in

The Journal of Biological Chemistry

Link to publication

Citation for published version (APA):

Montijn, R. C., van Rinsum, J., van Schagen, F. A., & Klis, F. M. (1994). Glucomannoproteins

in the cell wall of Saccharomyces cerevisiae contain a novel type of carbohydrate side chain.

The Journal of Biological Chemistry, 269(30), 19338-19342.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)

and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open

content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please

let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material

inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter

to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You

will be contacted as soon as possible.

(2)

0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Glucomannoproteins in the Cell Wall

of

Saccharomyces cerevisiae

Contain a Novel Type of Carbohydrate Side Chain*

(Received for publication, October 25, 1993, and in revised form, May 24, 1994)

Roy C. MontijnS, Johanna van Rinsum, Frank A. van Schagen, and Frans M. Klis

From the Institute of Molecular Cell Biology, University of Amsterdam, BioCentrum Amsterdam, Kruislaan 318,

1098 SM Amsterdam. The Netherlands

Mannoproteins in the walls of

mnn9

cells of

Saccha-

romyces cerevisiae

were released by laminarinase, and

purified by concanavalin

A

affinity chromatography and

ion-exchange chromatography. Carbohydrate analysis

revealed that they contained N-acetylglucosamine,

man-

nose, and glucose. An antiserum raised against p(l-6)-

glucan reacted with four proteins with molecular

masses of 66,100,155, and

220

kDa, respectively. Recog-

nition by the antiserum was competitively inhibited by

P(14)-glucan, but not by p(lS)-glucan, mannan, or dex-

tran

(an

a(l-6)-glucan). Mild periodate treatment of

the

wall proteins completely abolished recognition by the

antiserum. Glucose-containing

side chains were isolated

and compared with N- and O-carbohydrate side chains.

The

glucose-containing side chains consisted of about

equal amounts of

glucose and mannose and some

N-acetylglucosamine, and were larger than N-chains.

They were, however, not

extended N-chains, because af-

ter acetolysis, which preferentially cleaves (1-6)-1ink-

ages, their elution profiles differed strongly.

A

model is

presented of how glucose-containing side chains might

anchor mannoproteins into the glucan layer of the cell

wall.

Surface proteins in eukaryotic cells undergo various types of covalent modifications with glycans. Proteins that have en- tered the secretory pathway can be modified by the attachment of N-glycans to asparagine, and of O-glycans to serine or threonine.

In

yeast, the N-linked glycans consist of a chitobiose unit linked to asparagine and elongated

with

up to 150-200 mannose residues

(l),

and the O-linked glycans consist of one to five mannose residues (2, 3, 4). More recently, the attachment of a glycosylphosphatidylinositol anchor to the

C

terminus of specific membrane proteins has been demonstrated. Muller et

al.

(5)

showed

that

in yeast

the

glycosylphosphatidylinositol anchor of

a

plasma membrane-bound CAMP-binding protein contained glucosamine, mannose, and galactose residues.

Cell wall proteins of Saccharomyces cereuisiae can be divided into

two

groups. Some proteins can be extracted

with SDS,

whereas the remaining mannoproteins can be released only after digestion of the wall with a P(l-S)-glucanase, indicating that they are intimately associated with cell wall glucan

(6).

Recently, we found that glucanase-extractable wall proteins not only carry

N -

and O-linked side chains, but also glucose-containing side chains (7). Glucose was P-glycosidically linked, showing that the glucose-containing side chains did not represent incompletely processed N-linked side chains. We show

*

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Fax: 20-5257934.

$To whom correspondence should be addressed. Tel: 20-5257843;

here both immunologically and biochemically

that

the glucose- containing side chains are predominantly composed of ~(1-6)- linked mannose and p(l-61-linked glucose residues and represent a unique type of carbohydrate chain. We present a model that explains the contribution of these side chains to

the

attachment of mannoproteins to cell

wall

glucan.

EXPERIMENTAL PROCEDURES Materials

Mannose, mannitol, phenylmethylsulfonyl fluoride, BSA,' jack bean a-mannosidase, and mollusc laminarinase were purchased from Sigma. Laminarin (p(1-3)-glucan) and p-mercaptoethanol were purchased from Fluka AG. Pustulan (p(14)-glucan) was obtained from Calbio- chem. Bio-Gel P-10 and P-300 and prestained molecular weight stand- ards for Western analysis were from Bio-Rad. Concanavalin A (ConAI- Sepharose and DEAE-Trisacryl were obtained from Pharmacia Biotech Inc. Endo-p-N-acetylglucosaminidase H, peptide N-glycosidase F, and sweet almond P-glucosidase were purchased from Boehringer Mann- heim. Anhydrous hydrazine, BCA-protein assay reagent, and goat-anti- rabbit IgGhorseradish peroxidase were obtained from Pierce. All other chemicals were of analytical grade.

Methods

Yeast Strain and G r o w t h S . cerevisiae LB347-1C (mnn9, MATa) was kindly made available by Dr L. Ballou (Department of Biochemis- try, University of California, Berkeley, CA). Cells were grown at 28 "C in YPD medium (1% (w/v) yeast extract (Life Technologies, Inc.), 1% (w/v) Bacto-peptone (Difco), and 3% (w/v) glucose).

Isolation and Purification of Glucanase-extractable Proteins-Cell walls isolated from early exponential-phase cells (7) were boiled in 2% (w/v) SDS, 5 mM dithiothreitol, 10 mM Tris-HC1, pH 7.8, t o remove SDS-extractable mannoproteins. SDS-extracted cell walls were washed six times with 0.1 M sodium acetate, pH 5.5, containing 1 mM phenylmethylsulfonyl fluoride. To isolate glucanase-extractable mannoproteins, washed cell walls were resuspended in the same buffer (1 g in 2 ml), and 0.25 unit of mollusc laminarinase was added. After incubation at 35 "C for 2 h, again 0.25 unit of the enzyme was added followed by an additional incubation of 2 h. The glucanase-extractable mannoproteins were purified by ConA-Sepharose affinity chromatography and DEAE- Trisacryl anion-exchange chromatography (7). Part of the proteins were digested with Endo H for SDS-polyacrylamide gel electrophoresis and Western analysis (Scheme 1, Fraction I ) . Cell wall proteins, dissolved (8 mg/ml) in water containing 0.2% SDS (w/v) and 100 mM p-mercaptoethanol, were boiled for 5 min. SDS was removed by precipitating the glycoproteins in 9 volumes of cold acetone at -20 "C for 2 h. The pre- cipitate was evaporated and dissolved in 1 ml of 50 mM KH,PO,, pH 5.5, containing 100 mM p-mercaptoethanol and 0.5 mM phenylmethylsulfo- nyl fluoride. De-N-glycosylation was performed by adding 40 milliunits of Endo H. After incubating at 37 "C for 48 h, again 20 milliunits of Endo H were added to the mixture, and the incubation was continued for another 24 h.

Preparation of Neoglycoproteins-Pustulan (a p(1-6bglucose polymer with an average degree of polymerization of 120) was partly hydrolyzed in 0.1 M trifluoroacetic acid at 100 "C for 60 min to obtain 'The abbreviations used are: BSA, bovine serum albumin; C o d , concanavalin A , Endo H, endo-p-N-acetylglucosaminidase H; HPAEC, high pH anion-exchange chromatography; PAD, pulsed amperometric detection; ELISA, enzyme-linked immunosorbent assay; PBS, phos- phate-buffered saline.

19338

at Universiteit van Amsterdam on November 29, 2006

www.jbc.org

(3)

Glucomannoproteins in Yeast

19339

I

glass Lads homo enization with

I

-SDSextraction -1aminarinase digestion

-ConA-Sepharose -DEAE-Trisacryl

+I-

End/\ -PNGaseF -pelimination -hydrazinolysis -gel filtration

-gel filtration

J

_{t ' .}

Fraction I Fraction I1 Fractiw III

cell wall N-chains

p'",;;?"n

proteins

SCHEME 1. Flow diagram of the isolation of cell wall proteins and their

N-

and glucomannan chains.

water-soluble

p(

1-6)-glucan fragments. The hydrolysate was fraction- ated by gel filtration on Bio-Gel P-10. Fractions with a n average degree of polymerization of 15 hexose residues (as judged by gel filtration of peptide N-glycosidase F-released N-chains of mnn9 cell wall proteins) were pooled and lyophilized. Laminarin (a P(1-3)-glucose polymer with a n average degree of polymerization of 25), was oxidized in 0.25 M NaIO, a t 20 "C for 60 min. The reaction was stopped with ethylene glycol, and the oxidized laminarin was desalted and lyophilized (8). Both glucan preparations were conjugated to bovine serum albumin by reductive amination (9). After 5 days the reaction was stopped and the neoglycoproteins were isolated on a Bio-Gel P-300 gel filtration column. Both glucan-BSA conjugates contained approximately 30% carbohydrate. The p(l-G)-glucan-BSA conjugate was used for raising antibodies in rabbits. The p(1-3)-glucan-BSA conjugate was used for testing the specificity of the serum.

Characterization of Polyclonal Antibodies Specific for

p ( l - 6 -

Glucan-The binding specificity of the antibodies was determined using a combination of indirect ELISA against p(1-3)-glucan-BSA and

p(1-

6)-glucan-BSA, and indirect competitive ELISA using pustulan (p(1-6)- glucan), laminarin (p(l-3)-glucan), and yeast mannan. For the indirect ELISA, microtiter plates were coated with either p(1-3)-glucan-BSA or P(1-6)-glucan-BSA (100 p1 of 10 pg/ml in PBS, pH 7.2). The coated plates were washed four times with PBS and incubated with serial dilutions (up to 500,000) of antibodies diluted in PBS, 3% BSA, for 1 h a t 37 "C. The plates were washed four times with PBS and incubated with a goat-anti-rabbit IgG-peroxidase conjugate. After 1 h at 37 "C, the plates were washed as before and developed with a solution of tetra- methylbenzidine as chromogenic substrate (10). The reaction was al- lowed to progress at room temperature for 20 min and was stopped by the addition of 0.1 ml 1 M H,SO,. The plates were read at 450 nm. For the indirect competitive ELISA, microtiter plates were coated with P(l-B)-glucan-BSA (0.1 ml of 10 pg/ml i n PBS, pH 7.2). The coated plates were washed four times with PBS and incubated with p(1-6)- glucan antiserum diluted with PBS, 3% BSA (1/25,000), together with

the inhibitors a t t h e desired concentrations. The plates were developed as described.

Immunoblotting-Electrophoresis was performed on linear gradient (2.2-20%) polyacrylamide gels according to Laemmli (11). Proteins were either stained by the silver staining method as described by De Nobel et al. (12) or electrophoretically transferred to a n Immobilon polyvi- nylidne-&fluoride membrane for Western analysis. The membranes were blocked with 3% BSA in PBS and incubated with antibodies diluted with PBS, 3% BSA (1/25,000). Binding of the antibodies was detected with goat-anti-rabbit IgG-peroxidase using a solution of ami- noethylcarbazole as chromogenic substrate (10). The reaction was stopped by placing the blot in 50% ethanol. For competitive Western analysis, the proteins were blotted, the blot was blocked with 3% BSA in PBS, washed with PBS, and incubated with diluted p(1-6bglucan antiserum (1/25,000) together with 1 mM (glucose equivalents) of either pustulan, laminarin, mannan, or dextran (a(l-G)-glucan). For the oxi- dation of (14)-linkages of the glycan part of the proteins, the proteins were incubated with 50 mM periodic acid in 0.1 M sodium acetate, pH 4.5 for 1 h prior to incubation with the p(1-61-glucan antiserum.

Isolation a n d Analysis of Carbohydrate Side Chains from Glucanase- extractable Proteins-N-linked side chains were released by digestion with peptide N-glycosidase F (Fraction 11). The glucose-containing side chains were released together with the N-chains by hydrazinolysis (13) of glucanase-extractable proteins, which had earlier been freed from 0-chains by p-elimination (7). The liberated side chains were separated by size into the glucomannan and mannan fractions using Bio-Gel P-10 gel filtration as described in earlier work (Fraction 111) (7). Acetolysis was performed according to Ballou (1). Incubations with jack bean a-mannosidase (1 milliunit/mmol of hexose) and sweet almond p-glucosidase (1 milliunit/mmol of hexose) were performed in 0.1 M sodium acetate, pH 4.5, a t 37 "C for 24 h.

High pH Anion-exchange Chromatography with Pulsed Amperomet- ric Detection (HPAEC-PAD)-Separation of carbohydrates was performed on a CarboPac PA1 anion-exchange column (4 x 250 mm, Di- onex, Sunnyvale, CA), equipped with a CarboPac PA guard column ( 3 x 25 mm, Dionex). Carbohydrates were detected with the pulsed amperometric detector PAD I1 with a gold electrode (Dionex). To determine the carbohydrate composition, desalted fractions were hydrolyzed in trifluoroacetic acid (7). Monosaccharides were eluted with 15 mM NaOH. Oligosaccharides were eluted by a three-step procedure consisting of (a)

0.1 M NaOH for 10 min, (b) a linear gradient of 0-0.1 M sodium acetate in 0.1 M NaOH for 30 min, and ( c ) a linear gradient of 0.1-0.4 M sodium acetate in 0.1 M NaOH for 20 min. After every run the column was re-equilibrated in 0.1 M NaOH for 15 min. Chromatographic data were collected and analyzed using Dynamax software (Rainin, Emeryville, CA). Reference manno-oligosaccharides were generous gifts of Dr. K. HPrd, University of Utrecht, The Netherlands (Manal-BMan, Manal- 2Manal-2Man, and Mancrl-3Manal-2Manal-2Man), Dr. C. E. Bal-

lou, University of California, Berkeley (Mana1-2Man, Manal-GMan, Manal-2-Manal-2Man, Manal-GManal-GMan, Manal-2Manal-

2Manal-BMan, and Mancul3Manal-2Manal-2Man) and Dr. K. Ogawa, Iwaki Meisei University, Japan (Manal-BMan, Manal-3Man, Manal-GMan, Manal-2Manal-BMan, Manal-BManal-GMan, and Manal-GManal-6Man).

Analytical Methods-Protein concentrations were determined with the BCA-protein assay reagent with bovine serum albumin as a reference protein. Carbohydrate was measured with phenol-sulfuric acid with mannose as a reference (14).

RESULTS

Characterization of p(14)-Glucan Polyclonal Antibodies- For the detection and characterization of glucomannoproteins, polyclonal antibodies were raised against

p(

1-6)-glucan. Be- cause carbohydrates are weak immunogens in rabbits, we coupled a partial hydrolysate of pustulan (a P(1-6)-glucan), with an average degree of polymerization of 15 glucose resi- dues, to a carrier protein (BSA) to enhance the immune re- sponse. The binding specificity of the antiserum was deter- mined using indirect ELISA. Even at a 1/250,000 dilution, binding of the antiserum to P(1-6)-glucan-BSA could be de- tected (Fig. lA), whereas the cross-reactivity against p(1-3)- glucan-BSA dropped to zero at a serum dilution of 25,000. In the presence of 23 p pustulan (= 350 J ~ M glucose equivalents)

the binding to P(lL6)-glucan-BSA was halved.

No

inhibition by laminarin or mannan was observed in this assay. The results

www.jbc.org

(4)

*I

1 u - 1 u - 1 b 6 100

BI

60

1

PP

serum-dilution (l/x)

inhibitor

(pM)

FIG. 1. Characterization of the p(1-6)-glucan antiserum by indirect ELISA Panel A, binding of the antiserum to p(l-6)-glucan-BSA and P(1-3)-glucan-BSA. Closed circles, P(1-6)-glucan-BSA; open circles, P(1-3)-glucan-BSA. Panel B, competitive inhibition of the binding of the antiserum to P(1-6)-glucan-BSA by a pustulan hydrolysate (p(1-6)-glucan). Open circles, pustulan hydrolysate; closed circles, periodate-oxidized laminarin (P(13)-glucan); crosses, yeast mannan.

show that the serum has a strong specificity for p(1-6)-glucan.

Immunological Detection and Characterization

of

Glucosy-

luted Cell Wall Proteins-Glucanase-extractable proteins were

released from isolated walls by laminarinase (p(13)-glu-

canase) digestion and purified by Cod-Sepharose affinity

chromatography and DEAE-Trisacryl anion-exchange chroma-

tography. Mnn9 cells, which carry truncated N-chains ( E ) ,

were used for two reasons:

(i)

to obtain discrete protein bands

for SDS-PAGE and for Western analysis, and (ii)

to facilitate

the purification of glucose-containing chains (7). When purified

mannoproteins (Fraction

I) were separated by SDS-PAGE and

silver-stained (Fig.

2A),

several proteins were found with mo-

lecular masses of 410, 280, 220, 170, 145, 110, 66, 52, and 30

kDa, respectively.

After

Endo H digestion, proteins with mo-

lecular masses of 390, 250, 195, 135, 110,

and 52 kDa were

found, showing that most but not all proteins are sensitive

to

Endo H. In a Western analysis, the p(1-6)-glucan-specific an-

tibodies recognized four proteins with average molecular

masses of 220, 155, 100, and 66 kDa (Fig.

2 B ) .

These proteins

probably correspond with the 220-, 145, 110-, and 66-kDa

bands in the silver-stained gel (Fig.

2A).

Three of the four

proteins recognized by the anti-serum appeared to carry Endo

H-sensitive N-chains as shown by the reduction of the molecu-

lar masses from 220,155, and 100 kDa to 200,145, and 92 kDa,

respectively. The 66-kDa protein does not seem to carry Endo

H-sensitive N-chains. Table

I shows that in Endo H-digested

cell wall

mannoproteins the amounts of mannose and N-acetyl-

glucosamine had decreased due to the release of N-chains but

at the

same time the relative amount of glucose had increased

indicating that the glucose-containing side chains are insensi-

tive to Endo H.

To confirm that the epitope on these four cell wall proteins

indeed consisted of P(1-6)-glucan, they were first treated with

periodate.

As

shown in Fig. 3B, periodate treatment completely

abolished the recognition of the antibodies demonstrating the

carbohydrate nature of the epitope. Furthermore, the binding

of the antibodies could be inhibited competitively by partially

hydrolyzed pustulan (Fig. 3D), but not by mannan, laminarin,

or

dextran (Fig. 3, C, E , and F ) . These results demonstrate a

covalent association between four cell wall proteins and

p(1-

Acetolysis

of

N-chains and Glucose-containing Side Chains-

Both glucose-containing side chains (Fraction

111)

and N-chains

(Fraction

11)

contained mannose and N-acetylglucosamine al-

though in different ratios (7).

As

the glucose-containing side

chains were also considerably larger than normal N-chains (7)

(see also Fig. 4, A and B), an

obvious possibility was that they

actually represented modified N-chains obtained by the attach-

6)-glucan.

EndoH

-

+

" 170-

1 EndoH

-

+

66 -52

-

30

-

A

-

67 ₈₀

-

- 43

-

30

1

49.5

-

32.5

-

27.5

-

B

FIG. 2. SDS-polyacrylamide gel electrophoresis and Western analysis of glucanase-extractable wall proteins of mnn9 cells.

Proteins were released from isolated walls by laminarinase digestion and purified with ConA-Sepharose and DEAE-Trisarryl chromatography (Fraction I). For silver staining, 10 pg of protein were applied per lane. For Western analysis, 5 pg of protein were applied per lane. A, the gel was silver-stained. Lane I , before Endo H digestion; lane 2, after Endo H digestion. On the left the molecular masses of the proteins are indicated and on the right the molecular masses of the markers. The 30-kDa protein presumably represents ConA leached from the column during ConA-affinity chromatography. B, Western analysis of glu-

canase-extractable cell wall proteins using a P(l-6)-glucan antiserum. A serum dilution of V25,OOO was used. Lane 1, before Endo H digestion; lane 2, after Endo H digestion. The molecular masses of the prestained markers are indicated.

TABLE 1

Sugar composition of glucanase-extracted cell wall proteins in mnn9 cells (Fraction I)

Sepharose affinity chromatography and DEAE-Trisacryl anion-ex- The glucanase-extracted mannoproteins were purified by C o d - change chromatography. The carbohydrate composition of the purified proteins was determined by HPAEC-PAD.

Treatment Sugar

N-Acetylglucosamine Glucose Mannose

96

-

Endo H 2.7 4.6 92.7

+

Endo H 1.8 11.2 87.0

ment of P(1-6)-linked glucose residues to the core structure of

N-chains. To investigate this, N-chains and glucose-containing

chains were analyzed by acetolysis which selectively cleaves

(l-6)-linkages in oligo- and polysaccharides. Prior to acetolysis,

N-glycanase-released N-chains separated into five main peaks

www.jbc.org

(5)

control periodate

Glucomannoproteins in Yeast

mannan pustulan laminarin dextran

"

.

"F,'

A

B

C

D

E

F

FIG. 3. Characterization of the epitope of the /3(14)-glucan an- tiserum on glucanase-extractable cell wall proteins. Western analysis was performed as described in Fig. 2B. In each lane 5 pg of protein were applied. Lane A, control; lane B , after periodate treatment. Lanes C,

D,

E , and

F,

immunoblotting was carried out in the presence ofyeast mannan, pustulan hydrolysate (P(1-6)-glucan), laminarin

(p(1-

3)-glucan), and dextran (a(l-6)-glucan), respectively.

(Fig.

4A

1. It cannot be excluded

that this

heterogeneity is due

to

degradation occurring during digestion of the isolated walls

with laminarinase, since the enzyme preparation used con-

tained some a-mannosidase activity (7). Trimble and Atkinson

(16) have, however, shown that some heterogeneity occurs in

N-linked side chains of total cell mannoproteins. Hydrazinoly-

sis

released both N-linked chains and the glucose-containing

side chains. The glucose-containing side chains had a consid-

erably higher mass as shown by gel filtration and by HPAEC-

PAD

(compare Fig. 4,

A

and

B ) .

They were also heterogeneous

(Fig.

a).

Their composition was variable with an average mo-

lar ratio between mannose and glucose of 1:l. Fig. 4,

C

and

D ,

show the acetolysis products of purified N-chains and glucose-

containing side chains, respectively. Acetolysis of N-chains

yielded three peaks (Fig. 4C). Peak

Z

(15%) co-eluted under

separation conditions optimal for monosaccharides with man-

nose. Peak

ZZ

(59%) co-eluted under various separation condi-

tions with Mana(l-2)"an

and Mana(l3)"an.

Peak

IZZ

(18%) did not co-elute with any of the reference oligosaccha-

rides and presumably represents the chitobiose core

with 4 or 5

mannose residues attached to it (17). Acetolysis

of the glucose-

containing side chains resulted in a completely different pic-

ture (Fig.

40).

There was only one major monosaccharide peak

representing 72% of the total hexose and consisting of 43%

glucose and 57% mannose (peak

N).

The remaining minor

peaks did not co-elute with reference oligosaccharides under

various separation conditions and were not further character-

ized. Importantly, components similar to mannobioside(s) and

to chitobiose-mannosides, as found among the acetolysis prod-

ucts of normal N-chains (Fig. 4C,peaks

ZI

and

ZZZ),

were absent.

This strongly indicates that the glucose-containing side chain

are not derived from normal N-chains. Since acetolysis mainly

yielded monosaccharides, this shows that the majority of the

glucose and mannose residues in the glucose-containing side

chains are (14)-linked.

Enzymatic Degradation

of

Glucose-containing Side Chains-

Exo-a-mannosidase released 46% of the total hexose from the

purified glucomannan chains (Fraction

111) as monosaccharides

(Fig. 5). The released monosaccharide (Fig. 5, peak

Z)

co-eluted

with mannose under conditions optimal for separation of dif-

ferent monosaccharides (profile not shown). The release of 46%

of the total hexose as mannose by exo-a-mannosidase in com-

bination with the fact that glucose-containing side chains con-

12

345678

I 1 1 1 1 1 I I

A

I

I1

C

19341

400 200 0 400

g

0

>

200 v

5'

O E 400

%

z

3: 200 0 400 200 ' 0 1

Time

(min)

FIG. 4. Anion exchange chromatography (HPAEC-PAD) of N-

chains and glucose-containing side chains from glucanase-ex- tractable wall proteins before and after acetolysis. Panel A,

N-

linked side chains released by N-glycanase (Fraction 11) (66 nmol of hexose were injected and 42 nmol were recovered). Panel B , glucose- containing side chains released by hydrazinolysis (Fraction 111) (100 nmol of hexose were injected, and 78 nmol were recovered). Panel C, acetolysis products of N-linked side chains released by N-glycanase (60 nmol of hexose were acetolyzed, and 20 nmol were recovered as measured by HPAEC-PAD of the acetolysis products under optimal condi- tions for the separation of monosaccharides). Panel

D,

acetolysis products of glucose-containing side chains released by hydrazinolysis (100 nmol were acetolyzed, and 18 nmol were recovered by HPAEC-PAD of the acetolysis products under optimal conditions for the separation of

monosaccharides). Separation was performed by a three-step gradient consisting of 100 mM NaOH isocratic for 10 min followed by a linear gradient of 0-100 mM sodium acetate for 30 min and a linear gradient of 100-400 mM sodium acetate in 100 mM NaOH for 20 min a s indicated by the dotted line. The elution times of reference mono- and oligosac- charides are indicated by arrows: ( I ) mannitol, (2) Man, (3) Manal- GMan, ( 4 ) Manal-2Man and Manal3Man, ( 5 ) Manal-2-Manal-6- Man, (6) Manal-GManal-GMan and Manal-2Manal-2Man, (7)

Manal-2Manal-2Manal-2Man, (8) Manal-3Manal-2Manal- 2Man.

sist of approximately 50% mannose, indicates that the enzy-

matic degradation by exo-a-mannosidase was complete. These

results suggests that glucose-containing side chains consists of

a core of p(14)-glucan extended with one or more a(l-6)-man-

nose chains.

The results of exo-&glucosidase digestion

of purified glucose-

containing chains were difficult to interpret because the en-

zyme preparation was contaminated with a-mannosidase ac-

tivity capable of degrading N-chains (not shown). Indeed, the

released monosaccharides consisted of

4

4 %

glucose and 56%

mannose under separation conditions optimal for monosaccha-

www.jbc.org

(6)

Glucomannoproteins

in

I 11 1 1 1 1 1 1 12 3 4 5 6 7 8

Z

I 8

R

400

-

E

5' 200

f

0 10 20 30 40 50

60-

Time

(min)

chains (Fraction 111). (150 nmol were digested, and 87 nmol were

FIG. 5. a-Mannosidase digestion of glucose-containing side

recovered on HPAEC-PAD of the reaction products under optimal con-

were separated by HPAEC-PAD. The separation conditions were as ditions for the separation of monosaccharides). The reaction products described in Fig. 4. The arrows indicate the elution positions of the reference mono- and oligosaccharides.

rides. Even after prolonged digestion, glucose was never completely released indicating that part

of

the glucose residues were not accessible to the enzyme preparation.

DISCUSSION

The immunological data show that the cell wall of S. cerevi- siae

mnn9

cells contains four glucosylated proteins carrying a novel type of carbohydrate side chain characterized by the presence of p(l-6)-linked glucose residues. These proteins repre-

sent a specific subset of the total number of the glucanase- extractable proteins. Because we isolated these p(l-6)- glucosylated cell wall proteins after digestion of the cell wall with

p(

1-3)-glucanase, possible branching of this

p(

1-6)-glucan side chain with /3(1-3)-glucan would have been digested and would therefore not have been detected.

Glucose-containing chains cannot be released from cell wall glucomannoproteins by p-elimination (7) (data not shown) in-

dicating that they are not 0-linked to serine or threonine. They can, however, be freed by a harsher alkali treatment (7) or

hydrazinolysis (this paper). The similar molar ratio of glucose and mannose in the chains released by the two methods (l:l), as well

as

the fact that similar products were obtained by

chemical or enzymatic degradation, indicates that both meth- 10. Harlow, E.;and Lane, D. (1988)Antibodies:ALaboratory Manual. Cold Spring

ods released the same kind of chain. Acetolysis of glucose-con- taining side chains indicated that the majority of glucose and mannose residues are 1,6-linked and that only a few branch points are present. In combination with the results obtained by exo-a-mannosidase digestion, this indicates that most or all mannose residues are a( l-Ci)-linked. Similarly, exo-p-glucosidase digestion indicated that at least part of the glucose residues are P-linked. The limited release of glucose by p-glucosidase digestion might be due to the inaccessibility of glucose residues. Summarizing, the glucose-containing chains differ from known N-chains with respect to their resistance to peptide N-glycosidase

F

and Endo

H

(7), their composition, their be- havior on

HPAEC,

and with respect to the reaction products of chemical and enzymatic degradation. I t is therefore concluded

Harbor Laboratory, Cold Sprins Harbor, NY

that glucose-containing chains represent a novel type of carbo-

hydrate chain on cell wall mannoproteins. I t seems likely that at least part of the synthesis of the p(l-6)-glucan side chains takes place intracellularly.

It

is tempting to speculate that KRE5 and KREG are involved in the synthesis of this side chain, since both genes code for proteins involved in the synthesis of p(l-6)-glucan and are located in the endoplasmatic reticulum and Golgi, respectively (18).'

Although many structural features of yeast cell wall components are known, not much information is available on the architecture of the cell wall (for a recent review see Ref. 19). One of the major questions is how mannoproteins are anchored in the wall. The ability of P(l-S)-glucanase to liberate manno- proteins from cell walls suggests that they are closely associated with cell wall glucan

(4).

We suggest the following model for the anchoring of glucomannoproteins in the wall. According to this model, glucomannoproteins contain besides

N-

and

0-

linked chains, glucose-containing side chains linked to the protein moiety of the mannoprotein.

I n

vivo, the glucose-contain- ing side chains, which contain p(1-6)-linked glucose residues, might be extended with p(l-3)-linked glucose chains, which are removed during the isolation of the glucomannoproteins. These /3(1-3)-linked glucose chains might interweave with the

PCl-

3)-glucan fibrils in the cell wall thereby anchoring the glucomannoproteins.

Acknowledgments-We thank Dr. L. Ballou for kindly giving us the nnn9 mutant and Drs. K. HBrd, C. E. Ballou, and K. Ogawa for their generous gifts of manno-oligosaccharides. We also thank Herman van

den Ende and Heleen Caro for their comments on the manuscript. REFERENCES

1. Ballou, C. E. (1990) Methods Enzymol. 185, 440470

2. Zueco J.. Mormeneo. .S., .and Sentandreu. R. (1986) Bioehim. Biophys. Acta

&34,93-100

"

3. HBrd, K., Bitter, W., Kamerling J. P., and Vliegenthart J. E G . (1989) FEBS

4. Herscovics, A., and Orlean, P. (1993) J . FASEB 7, 540-550

5. Miiller, G., Schubert, K., Fiedler, F., and Bandlow, W. (1992) 1. Biol. Chem.

6. Valentin, E., Herrero, E., Pastor, E I. J., and Sentandreu, R. (1984) J. Gen.

7. Van Rinsum, J., a i s , F. M., and Van Den Ende, H. (1991) Yeast 7, 717-726 8. Meikle, P. J., Bonig, I., Hoogenraad, N. J., Clarke, A. E., and Stone, B. A. (1991)

Lett. 248, 111-114

267,25337-25346

Mic~obiol. 130, 1419-1428

Plnntn IRK. 1 4

. .~ ."

9. Roy, R., Katzenellenbogen, E., and Jennings, ---, - - J. J. (1984) Can. J. Biochem. Cell Biol. 62.270-275

11. Laemmli, U. K. (1970) Nature 227,680485

12. De Nobel, J. G., Dijkers, C . , Hooijberg, E., and a i s , F. Mi19891 J. Gen.

13. Takasaki, S., Mizuochi, T., and Kobata, A. (1982) Methods Enzymol. 83,263-

Microbiol. 135, 2077-2084 268

14. Dubois, M., Gilles, K. A., Hamilton, J. K., ReberE, P. A., and Smith, F. (1956) 15. Tsai, P. K , Frevert, J., and Ballou, C . E. (1984) J. Biol. Chem. 269,3805-3811 16. Trimble, R. B., and Atkinson, P. H. (1992) Glycobiology 2,57-75

17. Hernandez, L. M., Ballou, L., Alvarktdo, E., Gillece-Castro, B. L., Burlingame,

18. Meaden P., Hill, K. Wagner, J., Slipetz. D., Sommer, S. S., and Bussey, H.

19. Fleet, G . H. (1991) in The Yeasts (Rose, A. H., and Harrison, J. S., eds) Vol. 4,

Anal. Chem. 28,350-356

A. L., and Ballou, C. E. (1989) J. Biol. Chem. 264, 11849-11856 (1990) Mol. Cell. Biol. 10, 3013-3019

pp. 199-277, Academic Press, London

*

T. Roemer, personal communication.

www.jbc.org